Distributed virtual machines: a system architecture for network computing

Abstract

Modern virtual machines, such as Java and Inferno, are emerging as network computing platforms. While today’s virtual machines provide higher-level abstractions and more sophisticated services than their predecessors, and while they have migrated from dedicated mainframes to heterogeneous networked computers, their architecture has essentially remained intact. State of the art virtual machines are still monolithic, that is, all system components reside on the same host and are replicated among all clients in an organization. This crude replication of services among clients creates problems of security, manageability, performance and scalability. We propose a distributed architecture for virtual machines based on distributed service components. In our proposed system, services that control security, resource management, and code optimization are factored out of clients and reside in enterprisewide network servers. The services produce self-certifying, self-regulating, selfoptimizing programs via binary rewriting. We are currently building a Java virtual machine based on this architecture. We argue that distributed virtual machine architectures enable higher integrity, manageability, performance and scalability than monolithic virtual machines where all components reside on all clients.